In a recent issue of Schizophrenia Bulletin (June 2000, Vol. 26(2)), researchers from NIMH's Neuropsychiatry Branch, together with colleagues at Columbia University and the Kaiser Permanente Division of Research, published four papers from their observational Prenatal Determinants of Schizophrenia (PDS) study. The study group comprised 19,044 infants born between 1959 and 1967 in the prepaid Kaiser-Oakland Healthcare System. The pregnancies were carefully monitored with standardized instruments, and sera from each trimester were preserved, as well as cord blood. The unified records were computerized and the subjects continuously followed until they dropped out of the healthcare system. PDS tracked 12,576 of these subjects into adulthood. To date, 71 of these individuals have been identified as having a schizophrenia spectrum disorder. When possible, diagnosis was made through interviews with the subject. The investigators compared prenatal exposures for the individuals with schizophrenia spectrum disorder against appropriate controls from the same cohort. To date two findings have stood out. The mothers of the individuals who would eventually develop a schizophrenia spectrum disorder had: 1) an elevated prepregnancy body mass index (BMI), and 2) an excess of second trimester respiratory infections. BMI is a ratio of weight to height [(weight in pounds/height in inches)2 x 704.5]. Low BMIs are defined as less than or equal to 19.9; average is between 20.0 and 26.9; and high is considered to equal to or above 30.0. Having a high prepregnancy BMI increased the relative risk of schizophrenia from 1.0 for individuals with average BMIs to 2.9, or almost threefold. Furthermore, the risk of a schizophrenia spectrum disorder was double in the offspring of mothers who had a respiratory infection in the second trimester (risk of 2.06) compared with those who had an infection in the first or third trimester (risk below 1.0). These early antecedents of schizophrenia are presumed to exert their effects via a causal chain that may include prenatal as well as postnatal experiences. Body mass index, for example, may have many possible prenatal mediators, including poor maternal nutrition. Subsequent development of a schizophrenia spectrum disorder may also require the additional presence of a particular postnatal environment, and/or the involvement of genes. Respiratory infections during the second trimester of pregnancy have previously been identified as a risk factor for schizophrenia, but these previous studies have largely examined populations where there was an epidemic rather than knowing specifically which individuals had been infected (so called ecological studies). Future PDS studies will use the large amount of archived data, including sera, in an attempt to pin down some of the mediating factors involved in the development of schizophrenia and schizophrenia spectrum disorders. What is especially important about the findings from the PDS and other epidemiological studies is that they examine environmental influences that could potentially be controlled through public health interventions.